Guides
HomeChangelogStatus PageAPI ExplorerAPI ReferenceCLI ReferenceBlogRequest a Demo

Integrations

Suggest Edits

The following examples are queries related to tables, warehouses, and BI tools.

Please note that the sample queries below only show a subset of the fields available for brevity. You can use introspection to see all available fields.

Warehouses:

Get the Warehouse Id (dwId)

Your warehouse id can be retrieved using getUser.

The warehouse uuid (referred to as dwID) is used in subsequent queries to specify the warehouse context.

query getUser {
  getUser {
    account {
      warehouses {
        uuid
        connectionType
      }
    }
  }
}

{
  "data": {
    "getUser": {
      "account": {
        "warehouses": [
          {
            "uuid": "dummy_warehouse_id",
            "connectionType": "SNOWFLAKE"
          }
        ]
      }
    }
  }
}

Get Table Metadata

Table status and metadata can be retrieved using getTable. For multiple tables you can paginate through getTables instead.

The mcon can be used in subsequent queries as a unique identifier for the asset. If you already have the mcon that can be used in place of the fullTableId in the request.

query getTable {
  getTable(
    dwId:"dummy_warehouse_id", 
    fullTableId:"database:schema.table") {
      mcon
      fullTableId
      discoveredTime
      description
      location
      tableType
      createdTime
      status
      freshnessAnomaly
      sizeAnomaly
      freshnessSizeAnomaly
      metricAnomaly
      dynamicTable
      isDeleted
  }
}

{
  "data": {
    "getTable": {
      “mcon”: “dummy_table_id”
      "fullTableId": "database:schema.table",
      "discoveredTime": "2020-10-17T00:36:28.190057+00:00",
      "description": null,
      "location": null,
      "tableType": "TABLE",
      "createdTime": "2021-02-02T11:23:38.777000+00:00",
      "status": "Y",
      "freshnessAnomaly": false,
      "sizeAnomaly": false,
      "freshnessSizeAnomaly": false,
      "metricAnomaly": false,
      "dynamicTable": false,
      "isDeleted": false
    }
  }
}

Get Table Schema

The latest table schema (and previous versions of the schema) can be retrieved using getTable. These fields can be paginated through using the cursor.

query getTable {
  getTable(
    dwId:"dummy_warehouse_id", 
    fullTableId:"database:schema.table") {
      fullTableId
      schemaChangeCount
      versions (first:1) {
        edges {
          node {
            fields(first:10) {
              pageInfo {
                endCursor
                hasNextPage
              }
              edges {
                node {
                  name
                  fieldType
                }
              }
            }
          }
        }
      }
    }
}

{
  "data": {
    "getTable": {
      "fullTableId": "database:schema.table",
      "schemaChangeCount": 3,
      "versions": {
        "edges": [
          {
            "node": {
              "fields": {
                "pageInfo": {
                  "endCursor": ""
                  "hasNextPage": false
                },
                "edges": [
                  {
                    "node": {
                      "name": "ingestion_time",
                      "fieldType": "timestamp_ltz(9)"
                    }
                  }
                ]
              }
            }
          }
        ]
      }
    }
  }
}

Get Table Freshness

The table's freshness graph can be retrieved using getTable. 

query getTable {
  getTable(
    dwId:"dummy_warehouse_id", 
    fullTableId:"database:schema.table") {
      fullTableId
      mcon
      lastUpdates (startTime: "2021-03-24T06:59:59.999Z", endTime:"2021-03-31T06:59:59.999Z") {
        value
        measurementTimestamp
      }
    }
}

{
  "data": {
    "getTable": {
      "lastUpdates": [
        {
          "value": "2021-03-30T13:35:53+00:00",
          "measurementTimestamp": "2021-03-31T01:17:15+00:00"
        },
        {
          "value": "2021-03-30T01:34:16+00:00",
          "measurementTimestamp": "2021-03-30T13:07:03+00:00"
        },
        {
          "value": "2021-03-29T13:35:05+00:00",
          "measurementTimestamp": "2021-03-30T00:56:15+00:00"
        }
      ]
    }
  }
}

Get Table Size

The table's size (number of rows) can be retrieved using getTable. To query byte counts, replace totalRowCounts with totalByteCounts.

query getTable {
  getTable(
    dwId:"dummy_warehouse_id", 
    fullTableId:"database:schema.table") {
      totalRowCounts(startTime: "2021-03-24T06:59:59.999Z", endTime:"2021-03-31T06:59:59.999Z"){
        value
        measurementTimestamp
      }
    }
}

{
  "data": {
    "getTable": {
      "totalRowCounts": [
        {
          "value": 42,
          "measurementTimestamp": "2021-03-31T01:17:15+00:00"
        },
        {
          "value": 50,
          "measurementTimestamp": "2021-03-31T00:16:00+00:00"
        }
      ]
    }
  }
}

Get Table Tags

Table (or field) tags can be retrieved using getObjectProperties. Similarly, tags can be created and/or updated using createOrUpdateObjectProperty. These tags can be defined in the dashboard or retrieved from the warehouse. The results can be paginated through. 

query getObjectProperties{
  getObjectProperties(mconId:"dummy_table_id", first: 10) {
    edges {
      node {
        propertyName
        propertyValue
        propertySource
      }
    }
  }
}

{
  "data": {
    "getObjectProperties": {
      "edges": [
        {
          "node": {
            "propertyName": "is_important",
            "propertyValue": "true",
            "propertySource": SNOWFLAKE
          }
        }
      ]
    }
  }
}

📘

Using getTable

Please note that you can retrieve all these fields in one query (request). The separation was for demonstration purposes.

Get Connections of a Warehouse

Connections are also known as query-engines. Connections of a warehouse that support a specific job type can be retrieved using getWarehouseConnections:

query getWarehouseConnections {
  getWarehouseConnections(
    warehouseUuid: "dummy_warehouse_id"
    jobTypes: ["dummy_job_type"]
  ) {
    uuid
    name
  }
}

{
  "data": {
    "getObjectProperties": {
      "edges": [
        {
          "node": {
            "propertyName": "is_important",
            "propertyValue": "true",
            "propertySource": SNOWFLAKE
          }
        }
      ]
    }
  }
}

Add multiple query-engines

This section outlines a) how to add a nth query-engine within a single MC warehouse and b) how to add a custom monitor to the query-engine.

This allows creating custom monitors with multiple different connections. For instance, you can now leverage a larger Snowflake warehouse or higher priority Redshift WLM queue for certain field health or sql queries that are more resource-intensive without having to increase cost or load across the board. By default the dashboard and any APIs will use the primary (i.e. first) query-engine added to a warehouse.

📘

Note that you can always leverage the SDK for any of these operations!

Adding a new query-engine

Follow these steps to add a 2nd, 3rd or nth query-engine.

Step 1: Get the existing warehouse ID

Step 2: Test the new query-engine connection based on the the type of warehouse. Save the generated key. 

mutation testSnowflakeCredentials{
  testSnowflakeCredentials(account:"", password:"", user:"", warehouse:""){
    key
    success
    validations {
      type
      message
      data {
        database
        table
      }
    }
    warnings {
      type
      message
      data {
        database
        table
      }
    }
  }
}

Step 3: Save the connection

Add the query-engine connection you tested in step #2 to the warehouse from step #1. Save the generated UUID, it is necessary for any monitor APIs.

mutation addConnection {
	addConnection(dwId:"", connectionType:"", key:"", jobTypes:["sql_query", "json_schema"], connectionName: "friendly-connection-name") {
    connection {
      uuid
    }
  }
}

Options:

  • dwId - Warehouse UUID from step #1.
  • connectionType - Type of connection (e.g. snowflake for snowflake)
  • key - Key generated in step #2
  • jobTypes - Job types supported by the connection
  • connectionName - A friendly name for the connection

🚧

If the connectionType is a lake other than presto please remove json_schema from jobTypes. This feature is not supported and will result in an error.

Add a monitor using a nth engine

📘

Note that you can always leverage the SDK for any of these operations!

By default the dashboard and any APIs will use the primary (i.e. first) query-engine added to a warehouse. You can create a monitor for any nth engine by specifying the connectionId in a request.

The following monitor types are supported:

  • Field Health (createOrUpdateMonitor)
  • Dimension Tracking (createOrUpdateMonitor)
  • JSON Schema Changes (createOrUpdateMonitor)
  • SQL Rules (CreateOrUpdateCustomMetricRule)

📘

The term label or labels is used to refer to audiences in the API. labels and audiences are equivalent.

mutation createOrUpdateMonitor(
      mcon: $mcon
      monitorType: $monitorType
      fields: $fields
      timeAxisName: $timeAxisName
      timeAxisType: $timeAxisType
      scheduleConfig: $scheduleConfig
      uuid: $uuid
      whereCondition: $whereCondition
      aggTimeInterval: $aggTimeInterval
      lookbackDays: $lookbackDays
      segmentedExpressions: $segmentedExpressions
      notes: $notes
      fullTableId: $fullTableId
      resourceId: $resourceId
      description: $description
      labels: $labels # labels are equivalent to audiences
      connectionId: $connectionId
    ) {
      monitor {
        id
        uuid
        type
        fields
        entities
        timeAxisFieldName
        timeAxisFieldType
        aggTimeInterval
        aggSelectExpression
        historyDays
        whereCondition
        segmentedExpressions
        fullTableId
        notes
        selectExpressions {
          id
          expression
          dataType
          isRawColumnName
        }
        scheduleConfig {
          scheduleType
          intervalMinutes
          startTime
          minIntervalMinutes
        }
        schedule {
          resourceId
        }
      }
    }
  }

See below for example variables for the above call (note that labels refers to audiences):

{
  "mcon": "MCON++6cbdb767-f91b-462e-b906-81c8316d8265++0358854d-f853-45e2-9153-4f24b17ec1da++table++analytics:dbt_pc.recent_metrics",
  "monitorType": "stats",
  "fields": null,
  "timeAxisType": "timestamp_ltz(9)",
  "timeAxisName": "timestamp",
  "scheduleConfig": {
    "scheduleType": "FIXED",
    "intervalMinutes": 720
  },
  "aggTimeInterval": "HOUR",
  "lookbackDays": 1,
  "notes": "",
  "description": "Example FH with n-th engine",
  "labels": [],
  "connectionId": "a0499e90-b232-453f-9b22-fdaa3c1b0ded"
}

mutation createOrUpdateCustomMetricRule(
      comparisons: $comparisons
      customRuleUuid: $customRuleUuid
      customSql: $customSql
      variables: $variables
      description: $description
      dwId: $dwId
      intervalMinutes: $intervalMinutes
      startTime: $startTime
      timezone: $timezone
      notes: $notes
      scheduleConfig: $scheduleConfig
      labels: $labels # labels are equivalent to audiences
      connectionId: $connectionId
    ) {
      customRule {
        uuid
        ruleType
        intervalMinutes
        startTime
        timezone
        description
        creatorId
        nextExecutionTime
        createdTime
        updatedTime
        customSql
        variables
        comparisons {
          comparisonType
          fullTableId
          field
          metric
          operator
          threshold
          baselineAggFunction
          baselineIntervalMinutes
          isThresholdRelative
        }
        notes
      }
    }
  }

See below for example variables for the above call (note that labels refers to audiences):

{
  "comparisons": [
    {
      "comparisonType": "THRESHOLD",
      "operator": "GT",
      "threshold": 1
    }
  ],
  "customSql": "select * from analytics.dbt_pc.client_hub limit 10",
  "variables": "{}",
  "description": "Example custom sql rule with non-primary engine",
  "notes": "",
  "dwId": "0358854d-f853-45e2-9153-4f24b17ec1da",
  "connectionId":"a0499e90-b232-453f-9b22-fdaa3c1b0ded",
  "intervalMinutes": 10,
  "startTime": "2022-07-26T07:00:00.000Z",
  "timezone": "PDT",
  "labels": []
}

Limitations

  • The nth engine must be the same warehouse type as the primary (first) i.e. mixing warehouses is not currently supported.
  • Sampling or other any frontend requests / modifications will always leverage the primary (first) engine.

BI Tools:

Add Multiple LookML Repositories

In order to add more than one LookML repo, you will want to first to create an SSH Key. Next, you'll pass this SSH key into the below query.

Parameters:

repoUrl: Repository URL as ssh://[user@]server/project.git or the shorter form [user@]server:project.git
sshKey: SSH Key, base64-encoded

mutation {
  testLookerGitSshCredentials(
    repoUrl: "url_here"
    sshKey: "base64_encoded_ssh_key_here" 
  ){
    key
    success
  }
}

Once you have the returned key, you can pass this into the addBiConnection function.

Parameters:

connectionType: This will be looker-git-ssh in this case.
key: This is the key returned from the testLookerGitSshCredentials function

mutation{
  addBiConnection(
    connectionType: "looker-git-ssh"
    key: "your_key_here"
  ){
    connection{
      uuid
    }
  }
}

Updated 3 days ago